Proposed Fitness Function Research Articles

Effective TDMA scheduling for tree-based data collection using genetic algorithm in wireless sensor networks

Data collection is a major operation in Wireless Sensor Networks (WSNs) and minimizing the delay in transmitting the collected data is critical for a lot of applications where specific actions depend on the required deadline, such as event-based mission-critical applications. Scheduling algorithms such as Time Division Multiple Access (TDMA) are extensively used for data delivery with the aim of minimizing the time duration for transporting data to the sink. To minimize the average latency and the average normalized latency in TDMA, we propose a new efficient scheduling algorithm (ETDMA-GA) based on Genetic Algorithm(GA). ETDMA-GA minimizes the latency of communication where two dimensional encoding representations are designed to allocate slots and minimizes the total network latency using a proposed fitness function. The simulation results show that the performance of the proposed algorithm outperforms the existing state-of-the-art approaches such as Rand-LO, Depth-LO, DepthRe-LO, IDegRe-LO, and IDeg-LO in terms of average latency, average normalized latency, and average schedule length.

Choice of fitness functions and parameter settings in Genetic Algorithms for analysis of induction motors

The paper proposes methods for enhancing the accuracy and speed of Genetic Algorithms when determining the parameters of induction motors. A systematic study of the impact of genetic operators and algorithm parameters on optimization process performance is done. New fitness functions are developed containing additional quantities such as the input power of motors. Various genetic operator types and constraints are analysed providing an insight into a specific settings choice. Special attention is given to the factors generating stochastic noise and the ways to eliminate it. To evaluate the effectiveness of the proposed fitness functions and parameter settings, ten combinatorial optimization experiments are conducted on two types of induction motors. The results show that adequate fitness functions, combined with proper genetic operators setups, can significantly enhance the accuracy and speed of Genetic Algorithms while reducing the required input data.

Optimal design of IIR wideband digital differentiators and integrators using salp swarm algorithm

This paper proposes an approach based on a weighted L1-norm optimization criterion and employs it in conjunction with salp swarm algorithm (SSA) to design 2nd- to 4th-order wideband infinite impulse response (IIR) digital differentiators (DDs). Integration of the proposed fitness function and SSA allows both the magnitude and phase responses to be improved. An extensive simulation is carried out to explore the performance of the proposed approach. First, a comparative study with widely used methods, namely, real-coded genetic algorithm (RCGA) and particle swarm optimization (PSO), is performed to examine accuracy, robustness, consistency and efficiency. Afterwards, the magnitude and phase responses of the proposed digital differentiators are compared with those of the existing designs in the literature. New wideband IIR digital integrators (DIs) are derived by inverting transfer functions of their respective digital differentiators and then compared with the existing designs in the literature in terms of the magnitude and phase responses. Simulation and statistical results reveal that the superiority of the proposed approach is statistically significant and demonstrate that the proposed digital differentiators and integrators significantly outperform all state-of-the-art designs in terms of the magnitude response as measured by the absolute relative error (ARE) with almost linear phase responses in wideband frequency regions.

Efficient feature selection using one-pass generalized classifier neural network and binary bat algorithm with a novel fitness function

In high-dimensional data, many of the features are either irrelevant to the machine learning task or are redundant. These situations lead to two problems, firstly overfitting and secondly high computational overhead. The paper proposes a feature selection method to identify the relevant subset of features for the machine-learning task using wrapper approach. The wrapper approach uses the Binary Bat algorithm to select the set of features and One-pass Generalized Classifier Neural Network (OGCNN) to evaluate the selected set of features using a novel fitness function. The proposed fitness function accounts for the entropy of sensitivity and specificity along with accuracy of classifier and fraction of selected features. The fitness function is compared using four classifiers (Radial Basis Function Neural Network, Probabilistic Neural Network, Extreme Learning Machine and OGCNN) on six publicly available datasets. One-pass classifiers are chosen as these are computationally faster. The results suggest that OGCNN along with the novel fitness function performs well in the majority of cases.

Reliability enhancement of distribution networks using ESSs ancillary services: A probabilistic MILP methodology

This paper proposes a stochastic mixed integer linear programming (MILP) approach aiming at improvement of the reliability level of electric distribution network through a contingency-based energy storage systems (ESSs) incorporation, considering ESSs ancillary services and control sequences in service restoration. Further, charging stations installed in each ESSs are supposed to be purchased from various ESSs service providers with different charging/discharging rates. The proposed fitness function aims to simultaneously minimize the expected total cost of reliability (ETRC) and the expected system average interruption duration index (ESAIDI), to deal with economic and technical reliability aspects, respectively. The ETRC includes total cost of interruption and the total ESSs incorporation cost. In addition, the uncertainties in the restoration process in presence of the ESSs, including distribution system conditions and the available power of the cooperated ESSs during service restoration process are considered by several prospective scenarios. The effectiveness of the proposed methodology is investigated using a standard reliability test system (RBTS-4). The convergence of the proposed MILP approach is compared to the other methodologies presented in recent studies that proves the higher tractability of the proposed approach.

A placement architecture for a container as a service (CaaS) in a cloud environment

Unlike a traditional virtual machine (VM), a container is an emerging lightweight virtualization technology that operates at the operating system level to encapsulate a task and its library dependencies for execution. The Container as a Service (CaaS) strategy is gaining in popularity and is likely to become a prominent type of cloud service model. Placing container instances on virtual machine instances is a classical scheduling problem. Previous research has focused separately on either virtual machine placement on physical machines (PMs) or container, or only tasks without containerization, placement on virtual machines. However, this approach leads to underutilized or overutilized PMs as well as underutilized or overutilized VMs. Thus, there is a growing research interest in developing a container placement algorithm that considers the utilization of both instantiated VMs and used PMs simultaneously.The goal of this study is to improve resource utilization, in terms of number of CPU cores and memory size for both VMs and PMs, and to minimize the number of instantiated VMs and active PMs in a cloud environment. The proposed placement architecture employs scheduling heuristics, namely, Best Fit (BF) and Max Fit (MF), based on a fitness function that simultaneously evaluates the remaining resource waste of both PMs and VMs. In addition, another meta-heuristic placement algorithm is proposed that uses Ant Colony Optimization based on Best Fit (ACO-BF) with the proposed fitness function. Experimental results show that the proposed ACO-BF placement algorithm outperforms the BF and MF heuristics and maintains significant improvement of the resource utilization of both VMs and PMs.

Texture Image Classification Using Deep Neural Network and Binary Dragon Fly Optimization with a Novel Fitness Function

Texture is one of the most significant characteristics of an image for retrieving visually similar patterns. So far, researchers utilize large number of gray scale features and several combinations of training and testing. As a result, they do not guarantee high accuracy due to mismatch between gray scale features and classifier type. With a view to develop a highly accurate system, the paper proposes a new approach using Binary Dragon Fly with Deep Neural Network based fitness function for texture classification (BDADNN). Contributions of the proposed BDADNN is twofold, first fusion of Fractal, GLCM and GLRLM features has been performed to incorporate border complexities as well as spatial dependencies of a texture image, then Binary Dragon fly Algorithm is applied for feature selection with a novel fitness function based on Deep Neural Network (DNN).The proposed fitness function is designed such a way that it captures maximum accuracy, relevance and mainly minimizing number of features and reduction among features. Deep Neural Network is utilized as a stack of auto encoders. Adapting five types of k-cross-validation (k $$=$$ 2,3,4,5,6 and 10) protocols, we classify the texture images using BDADNN. Our results demonstrate superior performance of proposed BDADNN compared to SVM, for all cross-validation protocols (K2, K3, K4, K5 and K10) in terms of sensitivity, specificity and accuracy.

Shape optimization of preform tools in forging of aerofoil using a metamodel-assisted multi-island genetic algorithm

ABSTRACTPreform design plays an important role in improving the material flow, mechanical properties and reducing defects for forgings with complex shapes. In this paper, a study on shape optimization of preform tools in forging of an airfoil is carried out based on a multi-island genetic algorithm combined with a metamodel technique. An optimal Latin hypercube sampling technique is employed for sampling with the expected coverage of parameter space. Finite element (FE) simulations of multistep forging processes are implemented to obtain the objective function values for evaluating the forging qualities. For facilitating the optimization process, a radial basis function surrogate model is established to predict the responses of the hot forging process to the variation of the preform tool shape. In consideration of the compromise between different optimal objectives, a set of Pareto-optimal solutions are identified by the suggested genetic algorithm to provide more selections. Finally, according to the proposed fitness function, the best solution of multi-objective optimization on the Pareto front is confirmed and the corresponding preform tool shape proves optimal performances with substantially improved forging qualities via FE validation.

Robust beamforming and spatial precoding for quasi-OSTBC massive MIMO communications

This paper considers the problem of massive multiple-input-multiple-output (MIMO) wireless communication systems with quasi-orthogonal space-time block code (QOSTBC) transmission in the presence of spatial correlation effect (SCE) and mutual coupling effect (MCE). Conventional MIMO channels with QOSTBC transmission suffer from several drawbacks. The number of transmit antennas is restricted to a few in order to achieve full transmission rate. Therefore, it is difficult to make a quasi-orthogonal space time block coded massive MIMO system to achieve full transmission rate. Moreover, MIMO channel with QOSTBC transmission is usually considered for the case where the number of user equipments (UEs) is one. We present a joint beamforming and spatial precoding method to deal with the above drawbacks. The proposed method incorporates a beamforming scheme and spatial precoding to formulate an appropriate optimization process to effectively alleviate the considered problems. The resulting optimization problem will be solved by using a cooperative coevolutionary particle swarm optimization algorithm under two proposed fitness functions. During the optimization process, the proposed method finds the optimal beamforming coefficients of the precoding matrix, the optimal normalized positions of transmit and receive antenna elements for the case of using linear antenna arrays, and the optimal angle differences of transmit and receive antenna elements for the case of using circular arrays. Based on the proposed method, we are able to cure the performance degradation of a quasi-orthogonal space time block coded massive MIMO system due to the SCE and MCE. Moreover, the proposed method makes QOSTBC MIMO communications with full transmission rate for any number of transmit antennas achievable. Several simulation examples are presented to show the superior bit error rate (BER) performances of QOSTBC wireless MIMO scenarios with linear as well as circular antenna arrays by using the proposed method as compared to the existing methods.

A neuro-heuristic approach for recognition of lung diseases from X-ray images

Background and objectiveThe X-ray screening is one of the most popular methodologies in detection of respiratory system diseases. Chest organs are screened on the film or digital file which go to the doctor for evaluation. However, the analysis of x-ray images requires much experience and time. Clinical decision support is very important for medical examinations. The use of Computational Intelligence can simulate the evaluation and decision processes of a medical expert. We propose a method to provide a decision support for the doctor in order to help to consult each case faster and more precisely. MethodsWe use image descriptors based on the spatial distribution of Hue, Saturation and Brightness values in x-ray images, and a neural network co-working with heuristic algorithms (Moth-Flame, Ant Lion) to detect degenerated lung tissues in x-ray image. The neural network evaluates the image and if the possibility of a respiratory disease is detected, the heuristic method identifies the degenerated tissues in the x-ray image in detail based on the use of the proposed fitness function. ResultsThe average accuracy is 79.06% in pre-detection stage, similarly the sensitivity and the specificity averaged for three pre-classified diseases are 84.22% and 66.7%, respectively. The misclassification errors are 3.23% for false positives and 3.76% for false negatives. ConclusionsThe proposed neuro-heuristic approach addresses small changes in the structure of lung tissues, which appear in pneumonia, sarcoidosis or cancer and some consequences that may appear after the treatment. The results show high potential of the newly proposed method. Additionally, the method is flexible and has low computational burden.

Particle swarm Optimized Density-based Clustering and Classification: Supervised and unsupervised learning approaches

Abstract Two pattern recognition technologies in the field of machine learning, clustering and classification, have been applied in many domains. Density-based clustering is an essential clustering algorithm. The best known density-based clustering method is Density-Based Spatial Clustering of Applications with Noise (DBSCAN), which can find arbitrary shaped clusters in datasets. DBSCAN has three drawbacks: firstly, the parameters for DBSCAN are hard to set; secondly, the number of clusters cannot be controlled by the users; and thirdly, DBSCAN cannot directly be used as a classifier. In this paper a novel Particle swarm Optimized Density-based Clustering and Classification (PODCC) is proposed, designed to offset the drawbacks of DBSCAN. Particle Swarm Optimization (PSO), a widely used Evolutionary and Swarm Algorithm (ESA), has been applied in optimization problems in different research domains including data analytics. In PODCC, a variant of PSO, SPSO-2011, is used to search the parameter space so as to identify the best parameters for density-based clustering and classification. PODCC can function in terms of both Supervised and Unsupervised Learnings by applying the appropriate fitness functions proposed in this paper. With the proposed fitness function, users can set the number of clusters as input for PODCC. The proposed method was evaluated by testing ten synthetic datasets and ten benchmarking datasets selected from various open sources. The experimental results indicate that the proposed PODCC can perform better than some established methods, especially with respect to imbalanced datasets.

Hardware Trojan Detection Using an Advised Genetic Algorithm Based Logic Testing

Today, outsourced manufacturing of integrated circuit designs are prone to a range of malicious modifications of the circuitry called Hardware Trojans. HTs can alter the functionality of a circuit, leak secret information and initiate other possible malicious actions. HTs are activated in a very rare condition known by an intruder. Therefore, a group of HT detection methods tries to activate the HT circuitry by crafting test vectors. In this paper, we propose a logic testing based HT detection method using an advised genetic algorithm which creates effective test vectors, the so-called TRIAGE (hardware TR ojan detectI on using an A dvised G enetic algorithm based logic tE sting). The key contribution of this paper is to present a proper fitness function for the genetic algorithm providing better evaluation of the test vectors. The controllability, observability and transition probability factors of rare nodes have been considered in the fitness function. Simulation results indicate 80% reduction in generation time for test sets (on average) as compared to the previous work. On the other hand, reduced generation time for test vectors has been associated with an increase in trigger coverage. The coverage of the TRIAGE method for very hard to trigger Trojans increases by about 23% due to high efficiency of the proposed fitness function for the genetic algorithm.

Aggregative context-aware fitness functions based on feature selection for evolutionary learning of characteristic graph patterns

We propose aggregative context-aware fitness functions based on feature selection for evolutionary learning of characteristic graph patterns. The proposed fitness functions estimate the fitness of a set of correlated individuals rather than the sum of fitness of the individuals, and specify the fitness of an individual as its contribution degree in the context of the set. We apply the proposed fitness functions to our evolutionary learning, based on Genetic Programming, for obtaining characteristic block-preserving outerplanar graph patterns and characteristic TTSP graph patterns from positive and negative graph data. We report some experimental results on our evolutionary learning of characteristic graph patterns, using the context-aware fitness functions.

A hybrid model of Internet of Things and cloud computing to manage big data in health services applications

Over the last decade, there has been an increasing interest in big data research, especially for health services applications. The adoption of the cloud computing and the Internet of Things (IoT) paradigm in the healthcare field can bring several opportunities to medical IT, and experts believe that it can significantly improve healthcare services and contribute to its continuous and systematic innovation in a big data environment such as Industry 4.0 applications. However, the required resources to manage such data in a cloud-IoT environment are still a big challenge. Accordingly, this paper proposes a new model to optimize virtual machines selection (VMs) in cloud-IoT health services applications to efficiently manage a big amount of data in integrated industry 4.0. Industry 4.0 applications require to process and analyze big data, which come from different sources such as sensor data, without human intervention. The proposed model aims to enhance the performance of the healthcare systems by reducing the stakeholders’ request execution time, optimizing the required storage of patients’ big data and providing a real-time data retrieval mechanism for those applications. The architecture of the proposed hybrid cloud-IoT consists of four main components: stakeholders’ devices, stakeholders’ requests (tasks), cloud broker and network administrator. To optimize the VMs selection, three different well-known optimizers (Genetic Algorithm (GA), Particle swarm optimizer (PSO) and Parallel Particle swarm optimization (PPSO) are used to build the proposed model. To calculate the execution time of stakeholders’ requests, the proposed fitness function is a composition of three important criteria which are CPU utilization, turn-around time and waiting time. A set of experiments were conducted to provide a comparative study between those three optimizers regarding the execution time, the data processing speed, and the system efficiency. The proposed model is tested against the state-of-the-art method to evaluate its effectiveness. The results show that the proposed model outperforms on the state-of-the-art models in total execution time the rate of 50%. Also, the system efficiency regarding real-time data retrieve is significantly improved by 5.2%.

A new penalty-based wrapper fitness function for feature subset selection with evolutionary algorithms

ABSTRACTFeature subset selection is an important preprocessing task for any real life data mining or pattern recognition problem. Evolutionary computational (EC) algorithms are popular as a search algorithm for feature subset selection. With the classification accuracy as the fitness function, the EC algorithms end up with feature subsets having considerably high recognition accuracy but the number of residual features also remain quite high. For high dimensional data, reduction of number of features is also very important to minimize computational cost of overall classification process. In this work, a wrapper fitness function composed of classification accuracy with another penalty term which penalizes for large number of features has been proposed. The proposed wrapper fitness function is used for feature subset evaluation and subsequent selection of optimal feature subset with several EC algorithms. The simulation experiments are done with several benchmark data sets having small to large number of features. The simulation results show that the proposed wrapper fitness function is efficient in reducing the number of features in the final selected feature subset without significant reduction of classification accuracy. The proposed fitness function has been shown to perform well for high-dimensional data sets with dimension up to 10,000.

Navigation control of mobile robot using interval type-2 neural fuzzy controller optimized by dynamic group differential evolution

This study developed and effectively implemented an efficient navigation control of a mobile robot in unknown environments. The proposed navigation control method consists of mode manager, wall-following mode, and towards-goal mode. The interval type-2 neural fuzzy controller optimized by the dynamic group differential evolution is exploited for reinforcement learning to develop an adaptive wall-following controller. The wall-following performance of the robot is evaluated by a proposed fitness function. The mode manager switches to the proper mode according to the relation between the mobile robot and the environment, and an escape mechanism is added to prevent the robot falling into the dead cycle. The experimental results of wall-following show that dynamic group differential evolution is superior to other methods. In addition, the navigation control results further show that the moving track of proposed model is better than other methods and it successfully completes the navigation control in unknown environments.

Design and performance analysis of quantitative feedback theory basedautomated robust controller : An application to uncertain autonomous windpower system

Use of a robust controller for handling the operational uncertainties has become imperative in real time. This paper presents a modified fitness function based automated robust controller with the aid of quantitative feedback theory (QFT) using Genetic algorithm (GA). A controller exhibiting the desired decreasing modular plot and descending phase response is devised. The addition of arctangent function as one of the fitness function term is the proposed modification that facilitates in capturing the ideal controller characteristics. The proposed controller is applied to extract maximum power from a permanent magnet synchronous generator based autonomous wind power system. The step by step design guidelines for the automated QFT robust controller is deliberated in detail. The performance evaluation is carried out for step change and stochastically varying wind speed. Finally, benchmarking of the proposed controller against those available in the literature is accomplished through extensive simulations and it will be shown that the maximum power extraction along with least electromagnetic torque oscillations are achieved with the proposed fitness function based automated QFT controller.

A Cloud Resource Allocation Strategy Based on Fitness Based Live Migration and Clustering

With the advent of cloud computing and its boom in recent years, resource management tends to gain interest. Cloud offers different types of services necessary to mankind, the advancement being the virtual machines as resources in Infrastructure as a service. Our proposed fitness function detects the Hotspot and Coldspot to manage the resources dynamically by performing load balancing and server consolidation which is achieved through migration. The resource usage is maximized through utilization of the resources in their idle periods by forming clusters. The proposed system is implemented with open-source cloud framework OpenNebula. The experimental results prove that proposed system maximizes the resource utilization even in their idle periods.

Feature selection using firefly algorithm in software defect prediction

Defects occurring in software product are a universal event. Prevention of these defects in the early stage needs more attention because early stage prevention and fixing requires less effort and lower cost. Software defect prediction (SDP) is necessary in the determination of software quality as well as reliability. Prediction of defects is relatively an original research area in software quality engineering. Coverage of key predictors and the kind of data to be collected along with defect prediction model role, the interdependence of defects and predictors can be recognized in software quality. Feature selection (FS) is one of the worthy preprocessing techniques for application that uses huge volumes of data. It is the process of selecting the probable minimal attribute which is expected to be represented in the set of actual attributes. This paper proposes, FS using firefly algorithm (FA) and classifiers like support vector machine (SVM), Naive Bayes (NB) as well as K-nearest neighbor (KNN) are used for classifying the features selected. The FS that make use of the FA is that new technique of evolutionary computation that has been inspired by the process of flash lighting of the fireflies. This can search quickly the feature space for an optimal or a near optimal feature subset for minimizing a certain function of fitness. This proposed fitness function has made use of the incorporation of both the accuracy of classification and the reduction of the size. The results of the experiment have shown that the FS using the FA can achieve a better accuracy of classification than that of the other methods.

Optimal capacitor placement in distorted distribution systems considering resonance constraint using multi‐swarm particle swarm optimisation algorithm

Optimal capacitor placement in distribution systems has been studied for many reasons such as the improvement of voltage profile, mitigation of total harmonic distortion, and reduction of power losses. In this study, a new approach is presented for capacitor placement in the presence of harmonic distortion along with a proposed resonance index (RI). In this approach, a sensitivity analysis is first applied to find candidate buses. After that, the size and location of capacitors are optimised by a proposed fitness function. The fitness function includes a fuzzy membership function of annual net benefit, maximum THD of voltage, maximum voltage deviation, and a resonance constraint. This procedure supports different load levels employing fixed and switched capacitors. Also, a new RI is presented to prevent resonance conditions. Finally, the approach is tested on an IEEE 18-bus and an IEEE 69-bus system using the multi-swarm particle swarm optimisation algorithm. The simulation results show the efficiency of the method in comparison with the other methods.